org.bytedeco.dnnl

Class primitive_attr

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.dnnl.dnnl_primitive_attr_handle

org.bytedeco.dnnl.primitive_attr

All Implemented Interfaces:

AutoCloseable

@Namespace(value="dnnl")
 @Properties(inherit=dnnl.class)
public class primitive_attr
extends dnnl_primitive_attr_handle

\endcond

Primitive attributes.

See Also:

dev_guide_attributes

Nested Class Summary

Nested classes/interfaces inherited from class org.bytedeco.javacpp.Pointer

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator, Pointer.ReferenceCounter

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
primitive_attr()
primitive_attr(dnnl_primitive_attr t)
primitive_attr(dnnl_primitive_attr t, boolean weak)
primitive_attr(long size) Native array allocator.
primitive_attr(Pointer p) Pointer cast constructor.
primitive_attr(primitive_attr arg0)

Method Summary

Modifier and Type	Method and Description
void	get_output_scales(int[] mask, float[] scales)
void	get_output_scales(IntBuffer mask, FloatBuffer scales)
void	get_output_scales(IntPointer mask, FloatPointer scales) Returns output scaling factors correspondence mask and values.
post_ops	get_post_ops() Returns post-ops previously set via set_post_ops().
void	get_scales(int arg, int[] mask, float[] scales)
void	get_scales(int arg, IntBuffer mask, FloatBuffer scales)
void	get_scales(int arg, IntPointer mask, FloatPointer scales) Returns scaling factors correspondence mask and values for a given memory argument.
dnnl.scratchpad_mode	get_scratchpad_mode() Returns the scratchpad mode.
void	get_zero_points(int arg, int[] mask, int[] zero_points)
void	get_zero_points(int arg, IntBuffer mask, IntBuffer zero_points)
void	get_zero_points(int arg, IntPointer mask, IntPointer zero_points) Returns zero points correspondence mask and values.
primitive_attr	getPointer(long i)
primitive_attr	position(long position)
void	set_output_scales(int mask, float[] scales)
void	set_output_scales(int mask, FloatBuffer scales)
void	set_output_scales(int mask, FloatPointer scales) Sets output scaling factors correspondence mask and values.
void	set_post_ops(post_ops ops) Sets post-ops.
void	set_rnn_data_qparams(float scale, float shift) Sets quantization scale and shift parameters for RNN data tensors.
void	set_rnn_weights_qparams(int mask, float[] scales)
void	set_rnn_weights_qparams(int mask, FloatBuffer scales)
void	set_rnn_weights_qparams(int mask, FloatPointer scales) Sets quantization scaling factors for RNN weights tensors.
void	set_scales(int arg, int mask, float[] scales)
void	set_scales(int arg, int mask, FloatBuffer scales)
void	set_scales(int arg, int mask, FloatPointer scales) Sets scaling factors for primitive operations for a given memory argument.
void	set_scratchpad_mode(dnnl.scratchpad_mode mode) Sets scratchpad mode.
void	set_scratchpad_mode(int mode)
void	set_zero_points(int arg, int mask, int[] zero_points)
void	set_zero_points(int arg, int mask, IntBuffer zero_points)
void	set_zero_points(int arg, int mask, IntPointer zero_points) Sets zero points for primitive operations for a given memory argument.

Methods inherited from class org.bytedeco.dnnl.dnnl_primitive_attr_handle

asBoolean, asDnnl_primitive_attr, equals, get, get, notEquals, put, reset, reset

Methods inherited from class org.bytedeco.javacpp.Pointer

address, asBuffer, asByteBuffer, availablePhysicalBytes, calloc, capacity, capacity, close, deallocate, deallocate, deallocateReferences, deallocator, deallocator, equals, fill, formatBytes, free, getPointer, getPointer, getPointer, hashCode, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetof, offsetof, parseBytes, physicalBytes, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, sizeof, toString, totalBytes, totalCount, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

primitive_attr

public primitive_attr()

primitive_attr

public primitive_attr(@Const @ByRef
                      primitive_attr arg0)

primitive_attr

public primitive_attr(dnnl_primitive_attr t,
                      @Cast(value="bool")
                      boolean weak)

primitive_attr

public primitive_attr(dnnl_primitive_attr t)

primitive_attr

public primitive_attr(Pointer p)

Pointer cast constructor. Invokes Pointer.Pointer(Pointer).

primitive_attr

public primitive_attr(long size)

Native array allocator. Access with Pointer.position(long).

Method Detail

position

public primitive_attr position(long position)

Overrides:

position in class dnnl_primitive_attr_handle

getPointer

public primitive_attr getPointer(long i)

Overrides:

getPointer in class dnnl_primitive_attr_handle

get_scratchpad_mode

public dnnl.scratchpad_mode get_scratchpad_mode()

Returns the scratchpad mode.

set_scratchpad_mode

public void set_scratchpad_mode(dnnl.scratchpad_mode mode)

Sets scratchpad mode.

Parameters:

mode - Specified scratchpad mode.

set_scratchpad_mode

public void set_scratchpad_mode(@Cast(value="dnnl::scratchpad_mode")
                                int mode)

get_output_scales

public void get_output_scales(@ByRef
                              IntPointer mask,
                              @StdVector
                              FloatPointer scales)

Returns output scaling factors correspondence mask and values.

Parameters:

mask - Scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales vector. The set i-th bit indicates that a dedicated output scaling factor is used for each index along that dimension. The mask value of 0 implies a common output scaling factor for the whole output tensor.

scales - Vector of output scaling factors.

get_output_scales

public void get_output_scales(@ByRef
                              IntBuffer mask,
                              @StdVector
                              FloatBuffer scales)

get_output_scales

public void get_output_scales(@ByRef
                              int[] mask,
                              @StdVector
                              float[] scales)

set_output_scales

public void set_output_scales(int mask,
                              @StdVector
                              FloatPointer scales)

Sets output scaling factors correspondence mask and values. Example usage:

      int mb = 32, oc = 32,
          oh = 14, ow = 14; // convolution output params
      // unique output scales per output channel
      vector<float> scales = { ... };
      int oc_dim = 1; // mb_dim = 0, channel_dim = 1, height_dim = 2, ...
 
      // construct a convolution descriptor
      dnnl::convolution::desc conv_d;
 
      dnnl::primitive_attr attr;
      attr.set_output_scales(attr, oc, 1 << oc_dim, scales);
 
      dnnl::primitive_desc conv_pd(conv_d, attr, engine);
  

\note The order of dimensions does not depend on how elements are laid out in memory. For example: - for a 2D CNN activations tensor the order is always (n, c) - for a 4D CNN activations tensor the order is always (n, c, h, w) - for a 5D CNN weights tensor the order is always (g, oc, ic, kh, kw)

Parameters:

mask - Defines the correspondence between the output tensor dimensions and the \p scales vector. The set i-th bit indicates that a dedicated scaling factor is used for each index along that dimension. Set the mask to 0 to use a common output scaling factor for the whole output tensor.

scales - Constant vector of output scaling factors. If the scaling factors are known at the time of this call, the following equality must hold: scales.size() = \prod\limits_{d \in mask} output.dims[d]. Violations can only be detected when the attributes are used to create a primitive descriptor. If the scaling factors are not known at the time of the call, this vector must contain a single #DNNL_RUNTIME_F32_VAL value and the output scaling factors must be passed at execution time as an argument with index #DNNL_ARG_ATTR_OUTPUT_SCALES.

set_output_scales

public void set_output_scales(int mask,
                              @StdVector
                              FloatBuffer scales)

set_output_scales

public void set_output_scales(int mask,
                              @StdVector
                              float[] scales)

get_scales

public void get_scales(int arg,
                       @ByRef
                       IntPointer mask,
                       @StdVector
                       FloatPointer scales)

Returns scaling factors correspondence mask and values for a given memory argument.

Parameters:

arg - Parameter argument index as passed to the primitive::execute() call.

mask - Scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales vector. The set i-th bit indicates that a dedicated scaling factor is used for each index along that dimension. Set the mask to 0 to use a common scaling factor for the whole output tensor.

scales - Output vector of scaling factors.

get_scales

public void get_scales(int arg,
                       @ByRef
                       IntBuffer mask,
                       @StdVector
                       FloatBuffer scales)

get_scales

public void get_scales(int arg,
                       @ByRef
                       int[] mask,
                       @StdVector
                       float[] scales)

set_scales

public void set_scales(int arg,
                       int mask,
                       @StdVector
                       FloatPointer scales)

Sets scaling factors for primitive operations for a given memory argument.

Parameters:

arg - Parameter argument index as passed to the primitive::execute() call.

mask - Scaling factors correspondence mask that defines the correspondence between the tensor dimensions and the \p scales vector. The set i-th bit indicates that a dedicated scaling factor is used for each index along that dimension. Set the mask to 0 to use a common scaling factor for the whole output tensor.

scales - Constant vector of scaling factors. The following equality must hold: scales.size() = \prod\limits_{d \in mask} argument.dims[d].

See Also:

dnnl_primitive_attr_set_scales, dnnl::primitive_attr::set_output_scales

set_scales

public void set_scales(int arg,
                       int mask,
                       @StdVector
                       FloatBuffer scales)

set_scales

public void set_scales(int arg,
                       int mask,
                       @StdVector
                       float[] scales)

get_zero_points

public void get_zero_points(int arg,
                            @ByRef
                            IntPointer mask,
                            @StdVector
                            IntPointer zero_points)

Returns zero points correspondence mask and values.

Parameters:

arg - Parameter argument index as passed to the primitive::execute() call.

mask - Zero points correspondence mask that defines the correspondence between the output tensor dimensions and the \p zero_points vector. The set i-th bit indicates that a dedicated zero point is used for each index along that dimension. Set the mask to 0 to use a common zero point for the whole output tensor.

zero_points - Output vector of zero points.

get_zero_points

public void get_zero_points(int arg,
                            @ByRef
                            IntBuffer mask,
                            @StdVector
                            IntBuffer zero_points)

get_zero_points

public void get_zero_points(int arg,
                            @ByRef
                            int[] mask,
                            @StdVector
                            int[] zero_points)

set_zero_points

public void set_zero_points(int arg,
                            int mask,
                            @StdVector
                            IntPointer zero_points)

Sets zero points for primitive operations for a given memory argument.

Parameters:

arg - Parameter argument index as passed to the primitive::execute() call.

mask - Zero point correspondence mask that defines the correspondence between the tensor dimensions and the \p zero_points vector. The set i-th bit indicates that a dedicated zero point is used for each index along that dimension. Set the mask to 0 to use a common zero point for the whole output tensor.

zero_points - Constant vector of zero points. If the zero points are known at the time of this call, the following equality must hold: zero\_points.size() = \prod\limits_{d \in mask} argument.dims[d]. If the zero points are not known at the time of the call, this vector must contain a single #DNNL_RUNTIME_F32_VAL value and the zero points must be passed at execution time as an argument with index #DNNL_ARG_ATTR_ZERO_POINTS.

See Also:

dnnl_primitive_attr_set_zero_points, dnnl::primitive_attr::set_output_scales

set_zero_points

public void set_zero_points(int arg,
                            int mask,
                            @StdVector
                            IntBuffer zero_points)

set_zero_points

public void set_zero_points(int arg,
                            int mask,
                            @StdVector
                            int[] zero_points)

get_post_ops

@Const
 @ByVal
public post_ops get_post_ops()

Returns post-ops previously set via set_post_ops().

Returns:

Post-ops.

set_post_ops

public void set_post_ops(@Const @ByVal
                         post_ops ops)

Sets post-ops. \note There is no way to check whether the post-ops would be supported by the target primitive. Any error will be reported by the respective primitive descriptor constructor.

Parameters:

ops - Post-ops object to copy post-ops from.

set_rnn_data_qparams

public void set_rnn_data_qparams(float scale,
                                 float shift)

Sets quantization scale and shift parameters for RNN data tensors. For performance reasons, the low-precision configuration of the RNN primitives expect input activations to have the unsigned 8-bit integer data type. The scale and shift parameters are used to quantize floating-point data to unsigned integer and must be passed to the RNN primitive using attributes. The quantization formula is scale * (data + shift). \note Quantization scale and shift are common for src_layer, src_iter, dst_iter, and dst_layer. Example usage:

      // RNN parameters
      int l = 2, t = 2, mb = 32, sic = 32, slc = 32, dic = 32, dlc = 32;
      // Activations quantization parameters
      float scale = 2.0f, shift = 0.5f;
 
      primitive_attr attr;
 
      // Set scale and shift for int8 quantization of activation
      attr.set_rnn_data_qparams(scale, shift);
 
      // Create and configure rnn op_desc
      vanilla_rnn_forward::desc rnn_d(/* arguments * /);
      vanilla_rnn_forward::primitive_desc rnn_d(rnn_d, attr, engine);
  

Parameters:

scale - The value to scale the data by.

shift - The value to shift the data by.

set_rnn_weights_qparams

public void set_rnn_weights_qparams(int mask,
                                    @StdVector
                                    FloatPointer scales)

Sets quantization scaling factors for RNN weights tensors. The low-precision configuration of the RNN primitives expect input weights to use the signed 8-bit integer data type. The scaling factors are used to quantize floating-point data to signed integer and must be passed to RNN primitives using attributes. \note The dimension order is always native and does not depend on the actual layout used. For example, five-dimensional weights always have (l, d, i, g, o) logical dimension ordering. \note Quantization scales are common for weights_layer and weights_iteration

Parameters:

mask - Scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales vector. The set i-th bit indicates that a dedicated scaling factor should be used each index along that dimension. Set the mask to 0 to use a common scaling factor for the whole output tensor.

scales - Constant vector of output scaling factors. The following equality must hold: scales.size() = \prod\limits_{d \in mask} weights.dims[d]. Violations can only be detected when the attributes are used to create a primitive descriptor.

set_rnn_weights_qparams

public void set_rnn_weights_qparams(int mask,
                                    @StdVector
                                    FloatBuffer scales)

set_rnn_weights_qparams

public void set_rnn_weights_qparams(int mask,
                                    @StdVector
                                    float[] scales)